MHC class I genes, which provide immune surveillance against intracellular pathogens, are dynamically regulated by hormonal control. In the thyroid, thyroid stimulating hormone (TSH) represses class I transcription while triggering the production of thyroglobulin and secretion of thyroid hormone which, in turn, stimulates class I gene transcription. Together, these two hormones generate a dynamic cycle of class I regulation. These findings led us to suggest that this dynamic regulation maintains a constant level of cell surface presentation of self antigens; failure to appropriately regulate MHC class I genes would lead to excessive presentation of self antigens, and contribute to the generation of autoimmune disease. Consistent with this hypothesis, we have shown that in experimental models of autoimmune systemic lupus erythematosus and blepharitis, animals that fail to express class I are resistant to disease. This resistance is not due to the failure to generate CD8+ T cells in the absence of class I, since CD8-/- animals are highly susceptible to disease. Rather, the resistance appears to be due the failure to express class I in the periphery. Thus, in adoptive transfer experiments, susceptibility to the SLE-like disease is determined by the class I status of the recipient, not the class I status of the donor spleen cells. To study the role of class I in the etiology of spontaneous autoimmune disease, we have examined the NZBxNZW murine model of SLE. We have found that in NZBxNZW mice, levels of class I expression increase significantly with age. MMI-treatment, which prevents experimental SLE, also reduces the severity and incidence of this spontaneous autoimmune disease. To directly determine the role of class I in this spontaneous disease, the b2-microglobulin knock-out gene was introduced into each of the parental strains. Development of disease was compared in the completely class I-deficient NZB X NZW F1 mice with heterozygous Class I+/- and Class I+/+ NZB X NZW F1 mice. Interestingly, whereas all strains heterozygous for the b2-microglobulin knock-out gene were resistant to the onset of autoimmune disease, relative to the class I +/+ F1 mice, the F1 mice completely deficient in class I expression were more susceptible. These findings are consistent with the interpretation that reduced class I reduces the level of presentation of self-antigens which is protective. However, in the absence of any class I, CD8+ T cells fails to develop which eliminates a protective suppressor cell population. Experiments to test this hypothesis will need to be performed.